8

One can perform regular ordinary generalized linear model in R using glm function that has it's own method for summary function and one can summary of the model in which output there are p-values for each variable. Depending on those p-values one can say which variables are statistically significance or not under specific confidence level.

My question is. Is there is the same functionality for cv.glmnet function from glmnet package? I know that after computation I can receive a table with coefficients coef(model, s="lambda.min") where some of them are not zero. So I assume (maybe wrongly) that those non-zero are statistically significance. Am I right? Is there any method that provides p-values or confidence intervals for those coefficients?

eipi10
  • 91,525
  • 24
  • 209
  • 285
Marcin
  • 7,834
  • 8
  • 52
  • 99
  • 2
    Read this: http://stackoverflow.com/questions/12937331/r-glmnet-output-to-glm/17725220#17725220 Recent work has been done on producing a significance test, but we are not ready for math formulated in 2013 yet (http://arxiv.org/abs/1301.7161)... – Vlo Apr 27 '15 at 20:45
  • What a great amount of fresh statistics! Thank You for this – Marcin Apr 28 '15 at 11:15
  • @Vlo - please post that as an answer!! – smci Jun 17 '15 at 00:04
  • I took the liberty of retitling that question **[How to get statistical summary information from glmnet model?](http://stackoverflow.com/questions/12937331/how-to-get-statistical-summary-information-from-glmnet-model)** since that seems to be the intent. If that change is accepted, then let's close this one as a duplicate of that. – smci Jun 17 '15 at 00:24

0 Answers0